We expect to spend a significant amount of time on session handling changes next FY. It's less effort to do two changes to a given component at the same time than to do them separately, because you only incur the overhead of QA, communication, analysis etc once. So, if we have to refactor session handling to solve one problem, we should be on the lookout for opportunities to solve other problems with it on the cheap.

To do that, we should review existing tasks which propose changes to session handling.